Artificial intelligence (AI) recruitment tools could actually help remove gender biases in the job hiring process, new research has found.
Professor Andreas Leibbrandt from the Department of Economics at Monash University investigated how AI job recruitment tools can advantage or disadvantage women and underrepresented groups in two field experiments, looking at the behaviours of both applicants and recruiters.
“People in minority groups have inferior market outcomes, they earn less, they have a harder time finding and keeping a job,” Professor Leibbrandt said.
“It’s important to understand why that is the case so that we can identify and remove the barriers.”
The first-of-its-kind study first focused on how applicants perceive AI job recruitment tools. This field experiment engaged more than 700 applicants for a web designer position. The applicants were then informed whether their application would be assessed by AI or by a human job recruiter.
According to the research, women were much more likely to complete the job application when they discovered AI would be involved in the recruitment process. Meanwhile, men were less likely to apply for the job upon knowing an AI recruitment tool would be used in the job hiring process.
The experiment led Professor Leibbrandt to conclude women believe AI tools in the hiring process reduce bias, while men believe it removes an advantage for them.
The second part of the research studied the behaviour of more than 500 tech recruiters, and compared their scoring of job applicants with an AI tool’s assessment. It revealed that the recruiters still hold certain gender biases.
“We found that when recruiters knew the applicant’s gender, they consistently scored women lower than men,” Professor Leibbrandt said.
“However, this bias completely disappeared when the applicant’s gender was hidden.”
When recruiters accessed the applicant’s gender alongside the score given by the AI recruitment tool, there was no gender difference in scoring.
“This finding shows us they use AI as an aid and anchor – it helps remove the gender bias in assessment,” Professor Leibbrandt said.
The Monash University researcher said while academics continue to raise awareness about gender biases in AI technology, there must be a greater focus on the human interaction with AI and the gender biases that can come from that.
“My research isn’t just about dismantling bias,” Professor Leibbrandt said, “it’s about building a future of work where everyone has the opportunity to thrive.”
According to a study by Oxford Economics and Cognizant, 90 per cent of jobs will be affected by generative AI by 2032. More specifically, between 2023 and 2032, the percentage of jobs with high exposure points to AI could increase sixfold, from eight per cent to 52 per cent. It’s women’s jobs that will be most affected by generative AI over the next few years